GPU memory usage using Matlab deep learning

13 ビュー (過去 30 日間)
Amirreza Mahbod
Amirreza Mahbod 2017 年 12 月 12 日
コメント済み: Amirreza Mahbod 2017 年 12 月 20 日
Hi, I am using Matlab for fine tuning pre-trained CNN. The code works without any problems. However, when I monitor my GPU memory usage lively (using both GPU-Z and nvidia-smi in command line), I notice it varies between 4-7 GB. My GPU memory is 8 GB (GTX1070). Have any idea what is the reason behind that? The total running time is around 30 minutes for my code. I also observe much less running time on Ubuntu operation system when I tried to run the similar code on Keras.

サインインしてこの質問に回答する。

回答 (1 件)

Joss Knight
Joss Knight 2017 年 12 月 19 日
While you are training a network it needs a lot of working memory. The deeper the network, the more memory it needs. Which network are you training?
  1 件のコメント
Amirreza Mahbod
Amirreza Mahbod 2017 年 12 月 20 日
I am tying na d training different networks such as ResNet, VGG16 and GoogleNet

サインインしてコメントする。

カテゴリ

Help Center および File Exchange で Parallel and Cloud についてさらに検索

タグ

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by